P. Emery

[Accepted Manuscript] Randomized Controlled Trial of Rituximab and cost-effectiveness analysis in treating fatigue and oral dryness in primary Sjogren's Syndrome.

To investigate whether rituximab, an anti–B cell therapy, improves symptoms of fatigue and oral dryness in patients with primary Sjögrens syndrome (SS).

SAT0242 How good are the eular sjögren’s syndrome disease activity index (ESSDAI), and EULAR sjögren’s syndrome patients reported index (ESSPRI) in predicting health status in primary sjögren’s syndrome?

Background Over the past 2 years, the EULAR Sjogren’s syndrome study group have developed 2 new instruments, ESSDAI and ESSPRI to measure systemic disease activity and overall symptom burden. The ESSPRI also generates an EULAR sicca score (ESS) which measures the overall symptom of dryness. These instruments are designed to be used as standardised outcome measures for clinical studies and trials. Therefore it is useful to investigate how well these instruments predict the health status of patients with primary Sjogren’s syndrome (PSS). EQ-5D is a generic instrument that measure health outcome, the value sets can be converted to Time Trade Off (TTO) values representing the time a patient would be willing to give up to be freed from a reduced health state. In this study, we examined the relationship between ESSDAI and ESSPRI and the TTO values derived from EQ-5D. Objectives To evaluate the relationship between the two new instruments for the assessment of PSS (ESSDAI and ESSPRI) and health status of PSS patients. Methods Data including ESSDAI, ESSPRI and EQ-5D were prospectively collected from 633 PSS patients who have participated in the UK PSS registry (UKPSSR) using a standardised pro forma as previously described (1). TTO values were derived from the UK reference data provided by EuroQoL (the developer of the EQ-5D instrument) which has been transformed so that the values range from -1 to 1, with 1 being the number of years in perfect health state, 0 being dead and negative values representing health states worse than being dead. The relationships between the derived TTO values based on the health state of the patients and ESSDAI, ESSPRI as well as ESS were determined. Results The mean±SD TTO value of the PSS cohort was 0.624±0.301, with a range of -0.239 to 1. There were statistically significant correlations between TTO and ESSDAI, ESSPRI and ESS; TTO values decreased with increased ESSDAI, ESSPRI and ESS values (p<0.001 for all three). The strength of correlation was strongest with ESSPRI (R=-0.64), followed by ESS (R=-0.29) and ESSDAI (R=-0.15). Conclusions The recently developed EULAR PSS outcome assessment tools, in particular the ESSPRI, are useful predictors of the health status of PSS patients. Other UKPSSR collaborators: Moots R, Chadravarty K, Gendi N, Hamburger J, Richards A, Rauz S, Mulherin D, Kitas GD, Lloyd M, Lawson C, Clunie G, Knight S, Symmons D, Carr A, Carrozzo M, Figuereido F, Macleod I, Tarn JR, Foggo H, Edgar S,Young-Min S, Field A, Kaye S, Mewar D, Akil M, Dasgupta B, Fedele S, Porter S, Li C, Hall F. References Ng WF et al, Rheumatology, 2011;50:32-9. Disclosure of Interest None Declared

FRI0448 Evaluating health status of 620 patients with primary sjÖgren’s syndrome using EQ-5D

Background EQ-5D is a standardised tool for measurement of health status and is an increasingly popular health-related quality of life instrument but has not been applied to patients with primary Sjögren’s syndrome (PSS). EQ-5D provides a simple descriptive profile, a single index value for health status and a visual analogue score (VAS). The key advantages of EQ-5D are that the instrument is preference-based, easy to complete and the value sets can be easily converted to Quality Adjusted Life Years (QALY) to aid cost-utility analysis. Objectives To evaluate the health status of a large cohort of patients with primary Sjogren’s syndrome in the UK using EQ-5D. Methods We evaluated the health status of 620 clinically well characterised PSS patients from the UK PSS registry (UKPSSR) who fulfil the American European Consensus Group classification criteria 2002. All data were collected prospectively using a standardised pro forma as previously described (1). Data were compared to the UK normative data provided by the EuroQoL. In addition, the relationship between the health status of PSS patients and various clinical and patient reported outcome measures of PSS were examined. Results The proportion of PSS patients reporting any problem in mobility, self-care, usual activities, pain/discomfort & anxiety/depression were 42.4, 16.9, 56.7, 81.1 & 49.6 (%) respectively compared to 5.4, 1.6, 7.9, 30.2 & 15.7 (%) in the general UK population. The mean±SD VAS score was 59.9±21.2, compared to 81.3±16.8 for the general UK population. Univariate correlation analysis showed that EQ-5D VAS correlated with many clinical features of PSS but most strongly with fatigue, depression and pain with R values >0.5. Among the laboratory measures, only IgG levels, paraproteins and C3 correlated with EQ-5D VAS. Hierarchical cluster analysis showed that depression and fatigue are the most important determinants of variations in health status in this PSS cohort. Conclusions To our knowledge, this is the first report on the health status of PSS patients using EQ-5D. PSS patients have significantly impaired health status compared to the UK general population. Depression and fatigue are the key determinants of health status in PSS. Our data adds to the growing body of evidence that effective management of fatigue is key to improving the health status of PSS patients. References Ng WF et al, Rheumatology, 2011;50:32-9. Other UKPSSR collaborators: Moots R, Chadravarty K, Gendi N, Hamburger J, Richards A, Rauz S, Mulherin D, Kitas GD, Lloyd M, Moore L, Lawson C, Clunie G, Knight S, Symmons D, Carr A, Carrozzo M, Figuereido F, Macleod I, Tarn J, Foggo H, Mann S, Young-Min S, Field A, Kaye S, Mewar D, Akil M, Dasgupta B, Fedele S, Porter S, Li C, Hall F. Disclosure of Interest None Declared

SAT0162 Treatment Target Status at 6 Months and Long-Term Outcomes at 5 Years: Analysis of Methotrexate-Naïve Patients with Rheumatoid Arthritis in the GO-BEFORE Trial

Background Management guidelines recommend patients (pts) with RA should be treated with the intent of reaching a clinical target of low disease activity or remission at 6 months. Objectives To compare long-term efficacy outcomes for MTX-naïve RA pts in the GO-BEFORE trial grouped by their treatment target status at week 24 using DAS28-CRP score (<2.6, 2.6 to ≤3.2, >3.2) and SDAI (≤3.3, >3.3 to ≤11, >11). Methods In GO-BEFORE, 637 MTX-naïve pts with active RA were randomized to placebo (PBO)+MTX, golimumab (GLM)100mg +PBO, GLM 50mg+MTX, or GLM 100mg+MTX. Most PBO+MTX pts crossed over to GLM 50mg+MTX at wk52. In this analysis, pts were grouped by their treatment target status at wk24 using DAS28-CRP score (<2.6, 2.6 to ≤3.2, >3.2) and SDAI (≤3.3, >3.3 to ≤11, >11). Efficacy outcomes at 5 years (wk256) were evaluated for these mutually exclusive groups using observed data, Results At wk24, 150 pts had DAS28-CRP <2.6, 85 had DAS28-CRP 2.6 to ≤3.2, and 368 had DAS28-CRP >3.2. Of these, 23%, 31%, and 31%, respectively, discontinued treatment before wk256; 3%, 2%, and 3%, respectively, were due to lack of efficacy. Greater proportions of patients treated with GLM+MTX than patients treated with PBO+MTX improved from having DAS28-CRP >3.2 at wk24 to DAS28-CRP <2.6 at wk 52. Among pts achieving treatment targets at wk24, the majority either maintained a DAS28-CRP <2.6 (80%) or improved to DAS28-CRP<2.6 (72%) at week 256. Over 50% of pts with DAS28-CRP >3.2 at wk24 achieved treatment targets at wk256. Pts with DAS28-CRP≤3.2 at wk 24 had less progression in vdH-S scores and lower HAQ, SJC, TJC, pain, and pt/physician global assessment of disease (GAD) scores at wk 256 vs. patients with DAS28-CRP >3.2 at wk 24 (Table). Also, pts with DAS28-CRP 2.6 to ≤3.2 at wk24 had higher TJC (but not SJC), pain, and pt/physician GAD scores at wk256 than pts with DAS28-CRP <2.6 at wk 24. HAQ and change in vdH-S at wk256 were not significantly different between pts who achieved either DAS28-CRP <2.6 or DAS28-CRP 2.6 to ≤3.2 at wk 24. Of note, CRP levels at wk256 were similar among the three groups. Similar results were observed using SDAI scores (Table). Conclusions In GO-BEFORE, the majority of patients who achieved DAS28-CRP or SDAI score treatment targets at wk24 maintained or had improvement in clinical response at 5 years. At wk 256, efficacy outcomes, including clinical, functional, and radiographic scores, but not CRP level, at wk256 were significantly better among pts who had achieved DAS28-CRP <2.6 at wk24 vs pts with DAS28-CRP >3.2 at wk24. More subjective outcomes (TJC, pain, and pt/physician GAD) were also better at wk 256 for pts with DAS28-CRP <2.6 at wk 24 vs pts with DAS28-CRP 2.6 to ≤3.2 at wk24. Disclosure of Interest P. Emery Grant/research support from: Janssen R&D, LLC., Consultant for: Janssen R&D, LLC., R. Fleischmann Grant/research support from: Janssen R&D, LLC., Consultant for: Janssen R&D, LLC., S. Xu Employee of: Janssen R&D, LLC., E. Hsia Employee of: Janssen R&D, LLC.

AB0240 DAS28(CRP) <3.2 Achieved with or without A TNF Inhibitor: Differences in Radiographic, Clinical, and Biomarker Outcomes between Placebo + MTX and Golimumab + MTX Treatment among Mtx-NaÏVe Patients with Rheumatoid Arthritis

Background Treatment to lowest disease activity possible is the recommended treatment target for patients (pts) with rheumatoid arthritis (RA). However, radiographic damage and inflammatory biomarkers may differ based on treatment received and despite achieving the clinical target. Elevated inflammatory biomarkers have been associated with a higher risk of radiographic progression and extra-articular co-morbidities such as atherosclerosis. Objectives To explore differences in radiographic progression, clinical efficacy, and inflammatory biomarker levels in MTX-naïve RA pts with DAS28(CRP) <3.2 at Wk52, after treatment with MTX only or with GLM+MTX. Methods In GO-BEFORE, a randomized, double-blind, PBO-controlled trial, 637 MTX-naïve RA pts were randomized to PBO+MTX (n=160), golimumab (GLM) 100 mg+PBO (n=159), GLM 50 mg+MTX (n=159), and GLM 100 mg + MTX (n=159). At Wk52, the proportions of radiographic nonprogressors (change in vdH-S ≤0) and ACR20, ACR50, and ACR70 responders were determined. Changes from baseline to Wk52 in ESR, CRP, ICAM-1, IL-6, IL-8, and MMP-3 levels were measured. Outcomes were compared between the MTX only and the combined GLM (50 and 100 mg) +MTX groups among pts who did and did not have DAS28(CRP) <3.2 at Wk52. A similar analysis was performed for pts who did and did not achieve DAS28(CRP) <2.6. Results Among pts with DAS28(CRP) <3.2, clinical responses were similar between treatment groups, however a greater proportion of GLM+MTX-treated pts were radiographic nonprogressors vs. MTX only (p<0.05). Significantly greater improvements from baseline were also noted in levels of ICAM-1, IL-6, and IL-8 in GLM+MTX-treated pts (table). Among non-DAS28(CRP) <3.2 pts, no statistical differences were observed between treatment groups in radiographic, clinical efficacy, or biomarker outcomes. Overall, DAS28(CRP) <3.2 pts had better outcomes compared to those not achieving this threshold, regardless of treatment received. In the DAS28(CRP) <2.6 analysis, greater improvements in biomarker levels were also observed for those attaining target on MTX+GLM; however, difference in radiographic nonprogression did not reach significance. Pts not achieving DAS28(CRP) <2.6, nevertheless, showed significant radiographic nonprogression when on GLM+MTX compared to MTX alone. Conclusions Patients who attain DAS28(CRP) <3.2 on GLM+MTX exhibit greater radiographic nonprogression and suppression of ICAM-1, IL-6, and IL-8 compared to those on MTX alone at 1 yr. The potential benefits, beyond clinical response, of achieving pre-specified treatment goals with a TNF inhibitor may deserve further evaluation. Disclosure of Interest G. Karpouzas Consultant for: Janssen, Speakers bureau: Janssen, P. Emery Grant/research support: Janssen, Consultant for: Janssen, R. Fleischmann Grant/research support: Janssen, Consultant for: Janssen, S. Xu Employee of: Janssen Research & Development, LLC., E. Hsia Employee of: Janssen Research & Development, LLC. DOI 10.1136/annrheumdis-2014-eular.3794

SAT0166 Serum cytokine and chemokine profiling in primary sjögren’s syndrome

Background Primary Sjögren’s syndrome (pSS) is a multisystem autoimmune disease characterised by salivary and lacrimal gland inflammation leading to glandular destruction. The pathogenesis of pSS remains unclear, and research into pSS is made more challenging due to the heterogeneous clinical phenotypes among pSS patients. Several serum cytokines and chemokines have been linked with pSS pathogenesis, but their role in the various clinical manifestations in pSS have not been fully explored1,2. Objectives In this study, we examine the serum samples of a large cohort of clinically well-characterised PSS patients3 in order to determine whether serum cytokines and chemokines may be used to differentiate PSS patients from healthy controls, and if so, the relationship between these serum abnormalities and clinical phenotypes. Methods Serum levels of 24 different cytokines, chemokines and adhesion molecules for 150 pSS patients and 30 healthy controls were measured using Cytometric Bead Array.PSS patients were further classified into the following subsets: 1) Lymphoma; 2) No Lymphoma; 3) High systemic disease activity (ESSDAI score >12); 4) Low systemic disease activity (ESSDAI <1); 5) High residual glandular function (Oral salivary flow (OSF) >10ml/15min and Schirmer’s test >10cm); 6) Low residual glandular function (OSF <1ml/15min and Schirmer’s test <1cm); 7) Anti-Ro and Anti-La positive; 8) Anti-Ro and Anti-La negative. The relationship between analyte levels and clinical and laboratory parameters of PSS was examined using multivariate analysis and Mann-Whitney U testing; p-values were Bonferroni corrected for multiple comparisons. Results There were marked differences in the levels of many cytokines and chemokines between PSS patients and healthy controls, with a p value <0.001, statistically significant after Bonforroni’s correction for multiple comparisons. Serum IL4 and IL17 were found to be significantly higher in patients versus controls. Chemokines such as MIG (CXCL9), MIP1a (CCL3) and MIP1b (CCL4), IP10 (CXCL10), were also measured at higher levels in patient serum. Serum levels of IFNa, LTα and TNFα also differ significantly between patients and controls. Differences in cytokine levels were observed between patient subsets which were no longer significant after Bonferroni correction. Conclusions Differences in blood cytokine and chemokine levels between primary Sjogren’s patients and controls can be detected in serum. Serum MIG (CXCL9), MIP1a (CCL3) and MIP1b (CCL4), IP10 (CXCL10) IFNa, LTA and TNFa levels differ significantly between patients and controls. Our observations raise the possibility that these analytes may be important in disease pathogenesis. References Baturone, R. et al. Scand. J. Immunol. 2009;38:386. Szodoray, P. et al. Scand. J. Immunol. 2004;59:592. Ng, W.F. et al. Rheumatology 2010;50:32. Disclosure of Interest None Declared